industrial investigations of the liquid steel filtration

339METALURGIJA 53 (2014) 3, 339-342

K. JANISZEWSKI, B. PANIC

INDUSTRIAL INVESTIGATIONS OF THE LIQUID STEEL FILTRATION

Received – Prispjelo: 2013-08-23Accepted – Prihvaćeno: 2013-12-30

Preliminary Note – Prethodno priopćenje

ISSN 0543-5846METABK 53(3) 339-342 (2014)

UDC – UDK 669.186:518.1:531.3=111

K. Janiszewski, B. Panic: Silesian University of Technology, Depart-ment of Metallurgy, Katowice, Poland

Hitherto existing investigations concerning the ceramic fi lter use in the steel making processes have given good results. The obtained results of fi ltration have proved that this method may be used as an eff ective and cheap way of steel fi ltration from non-metallic inclusions. Placing fi lters in the tundish is the best location considering the limi-tation of the possibility of secondary pollution of steel. Yet, the results presented in this paper, of an experiment prepared and carried out in the industrial environment, are the only positive results obtained, which are connected with so much quantities of liquid steel processed with use of the multi-hole ceramic fi lters.

Key words: steel, refi ning, tundish, ceramic fi lter, continuous casting (CC)

INTRODUCTION

It is evident from the hitherto existing experience [1, 2] that the conventional out-of-furnace steel treatment (espe-cially that which is deoxidized by depositing, e.g. with use of aluminium) does not ensure high levels of the metal-lurgical purity. Furthermore, the presence in liquid steel of non-metallic inclusions of Al2O3 type throws into confu-sion the process of continuous casting due to the phenom-enon of covering the ladle discharge nozzles by a layer of such inclusions. According to judgements presented by many research centres [3, 4-8] the steel fi ltration with use of multi-hole ceramic fi lters can be the effi cient and cost-effective method of removing the non-metallic inclusions from liquid steel. The experimental results obtained hith-erto in the laboratory and fi eld indicate the substantial re-duction in content of non-metallic inclusions and damag-ing impurities in liquid steel [4, 6, 8]. Differences, how-ever, exist in levels of effi ciency of this steel refi ning method, depending on local fi ltration conditions. The rea-son for such differences can be found in the phenomenon of secondary oxidation of liquid steel by the atmospheric oxygen [8]. The positive results obtained in the laboratory-scale research have bcecome the base to undertake the tri-als to fi ltrate liquid steel in industrial conditions. A series of model investigations has been carried out, and then, af-ter obtaining the positive results, a series of industrial-scale melts of steel has been produced [9–13]. The goal of the research carried out, the results of which are presented here, has been to prove the possible extent of the solid non-metallic inclusion removal from liquid steel through the steel fi ltration by means of multiple-orifi ce ceramic fi lters. The aim of the research carried out has been to prove that

the liquid steel fi ltration is a cheap and effi cient additional processing stage, separating the non-metallic inclusions, which in case of the conventional casting technology could remain in the cast steel.

INDUSTRIAL INVESTIGATIONS OF STEEL FILTRATION

The hitherto obtained results of the laboratory re-searches of liquid steel fi ltration by means of ceramic fi l-ters [4, 6, 8] have become the base for preparation and implementation of the industrial application of the steel fi ltration process in the processing line of the continuous casting machine. Figure 1 presents the trough-type tundish prepared, in which a ceramic fi lter has been mounted.

The fi lter used, in form of a barrier, has been made with 26 orifi ces of 60 mm diameter and fi ltrating surface of 808 236 mm2. The fi ltrating orifi ces have been of 165 mm in

Figure 1 Tundish of the CC machine provided with the ceramic fi lter

340 METALURGIJA 53 (2014) 3, 339-342

K. JANISZEWSKI et al.: INDUSTRIAL INVESTIGATIONS OF THE LIQUID STEEL FILTRATION

where:xp – inclusion surface share (or inclusion number) before

fi ltration,xk – inclusion surface share (or inclusion number) after fi l-

tration, with use of the following intervals of inclusion diame-

ters according to Ferret: 0,5-2,5 µm, 2,6-6,5 µm, 6,6-15 µm, 15,6-30 µm, what is shown in Figure 3. Figures 4–9 show the surface shares of the non-metallic inclusions in fi ltrated and non-fi ltrated steel for the experimental melts, corre-spondingly for each of the Fx Feret’s diameter intervals. The border of phase division between fi lter ceramics and solidi-fi ed steel together with adjoining areas (after polishing the sample surface and deposition of thin fi lm of gold) has been examined with the X-ray microanalyses method by means of Noran Instrument’s Hitachi S-3500N scanning micro-scope. The highest level of the effectiveness of liquid steel fi ltration has been observed for inclusions in the 2,6 – 30,0 µm interval, what has been confi rmed by the previously ob-tained experimental results [3, 5, 7].

The observed inclusions differ in shape and size. In the aluminium deoxidized melts (the second trial and the third trial) the single inclusions as well as irregular in shape the inclusions clusters of different confi guration have been ob-served. They are built from non-metallic phase (Al2O3)

length. The fi lter has been manufactured by Alcor S.A. company of Krzeszowice, Poland, and has been made of mullite-based body. In evaluating the effectiveness of the fi ltration process has benefi ted from the widely presented in the literature for fi lter design slenderness [6]:

PF

FF P

PS⋅

=544,3

(1)

where: PF – the fi ltrating surface of the fi lter channel / m2,PPF – the surface area of the fi lter channel cross-section

/ m2,SF – the fi lter slenderness ratios.

Ten (10) melts of A700 steel (rail steel), about 330 Mg each, have been in a sequential manner casted during the fi rst trial of industrial steel fi ltration. After termination of the steel casting sequence the „wash-out” effect has been dis-covered in the fi lter central part. The initial assessment of the macroscopic structure of the continuous castings of the fi ltrated steel (not including the investigation of steel con-tamination with non-metallic inclusions) has proved good product quality. The fi rst carried out industrial trial of steel fi ltration in the tundish of CC machine has not thrown the continuous casting process into any confusion and not proved the earlier apprehension of emergency risk. The sec-ond trial of industrial steel fi ltration, in the tundish of CC machine, comprise a sequence of three melts of SE03-u steel (chemical composition steel: C- 0,340, Mn- 0,840, Si- 0,250, P- 0,025, S- 0,020, Al- 0,045), about 330 Mg in weight each. During the fi ltration process the samples of fi l-trated steel have been collected from one half of the ladle, while from the second, non-fi ltrated one the samples have been collected for analysis of total oxygen content. Results of the investigations carried out are presented in graphical form in Figure 2. The oxygen content analysis has been car-ried out with use of the Leco company’s method. Effective-ness of steel fi ltration has been evaluated as variation in the surface share of the non-metallic inclusions in fi ltrated steel versus the non-fi ltrated one according to the formula (2):

%100⋅−

=p

kpN M I x

xxη (2)

Figure 2 (O)T total oxygen content variation in each of the melt sequence - 34GJ steel

Figure 3 The eff ectiveness of removing non-metallic inclusions as measured with the average rate of non-metallic inclusion superfi cial share ηNMI - 34GJ steel

Figure 4 The eff ectiveness of removing non-metallic inclusions as measured with the average rate of non-metallic inclusion superfi cial share ηNMI, with division into inclusion size intervals according to Fx Feret diameters melt no 1 - SE03-u steel

341METALURGIJA 53 (2014) 3, 339-342


produced during the deoxidizing process. They consist of Al2O3 non-metallic phase, being the product of the deoxi-dizing process. In every experimental melt (excluding in-clusions in M-3 melt - the third trial) the decrease of the inclusion surface-share in the fi ltrated steel in comparison with the non-fi ltrated steel was observed.

The highest degree in the surface-share decrease in re-lation to all inclusions was observed in M-1 - M-6 melts with parameters ηNMI equal to 2,20 %, 1,94 % -10,10 %, 31,10 %, 7,53 % and 27,21 % respectively (the second trial). Figure 3 illustrate this dependence for average ηNMI value calculated for every experimental melt. Increase of the non-metallic inclusions size (measured with diameter Fx) causes increase in the number of inclusions removed during fi ltration, measured with ηNMI.

SUMMARY AND CONCLUSIONS

Based on the research carried out hitherto and the pub-lished results, a judgement should be made that liquid steel fi ltration with the ceramic fi lters can become in the nearest future the effective and cheap method of additional steel

refi ning, in order to separate the non-metallic inclusions, as well as the permanent processing procedure in the con-tinuous casting technology (for some types of steel). The model research has proved good liquid steel fl ow dynam-

Figure 5 The eff ectiveness of removing non metallic inclusions as measured with the average rate of non-metallic inclusion superfi cial share ηNMI, with division into inclusion size intervals according to Fx Feret diameters melt no 2 – SE03-u steel





342 METALURGIJA 53 (2014) 3, 339-342


ics and steel mixing in the tundish provided with multiple-orifi ce fi lters in case of fi lter installation in the place of conventional overfl ow partitions. Placing the multiple-ori-fi ce fi lters in the position of conventional overfl ow parti-tions is the most benefi cial fi lter location within the way of steel making process. The fi ltration trials carried out have not proved the earlier apprehension of emergency risk, have not thrown into confusion the continuous casting process and even have improved the liquid steel mixing dynamics in the tundish, as it has become evident after the model research. The fi lter washout effect discovered in its lower part (the fi rst trial) should not be considered a sur-prise due to the fact that the quantity of liquid steel proc-essed in each melt is about 330 Mg. The ηWN effectiveness of non-metallic inclusions removal, as measured by the av-erage extent of variation in the inclusion surface share in fi ltrated and non-fi ltrated steel, has amounted 9,98 % for the second trial with the fi lter slenderness ratio SF = 3,10. There is a one more positive effect of use of the multiple-orifi ce ceramic fi lters, which is the increased time of the ladle nozzle service life. If part of the non-metallic inclu-sions is stopped at the level of the tundish then the nozzle service life is decidedly increased (the process of decrease in the nozzle cross section runs more slowly).

REFERENCES

[1] W. I. Jawojski and others: Wkljuczienia i gazy w staliach. Wyd. Mietałłurgia, Moskwa, 1979.

[2] J Happ, M.G. Fröhberg: Giessereiforschung, 1971, Heft 1, 1-9.

[3] K. Janiszewski: Refi ning of liquid steel in a tundish using the method of fi ltration during its casting in the CC machi-ne, Archives of Metallurgy and Materials, 58 (2013) 2, 509-517.

[4] L. Socha, J. Bažan, L. Martínek, P. Fila, M. Balcar, P. Lev: Laboratory Verifi cation of Resistance of Refractory Mate-

rials for Ceramic Filters, METAL 2010, Rožnov p. Radhoštěm, Czech Republic, 2010, 90-95.

[5] J. Mancini, J. Stel: Tundish metallurgy: a combined Irsid und Hoogovens research, Revue de Métallurgie-CIT, 89 (1992) 3, 269-277.

[6] K. Janiszewski, The slenderness ratio of the fi lter used in the process of liquid steel fi ltration as the additional para-meter of the fi lter form, Steel Research International, 84 (2013) 3, 288-296.

[7] K. Janiszewski: Industrial application of liquid steel fi ltra-tion out of dispersed nonmetallic phase in the continuous casting machine, Metalurgija 52 (2013) 1, 71-74.

[8] K. Janiszewski: Infl uence of slenderness ratios of a multi-hole ceramic fi lters at the effectiveness of process of fi ltra-tion of non-metallic inclusions from liquid steel, Archives of metallurgy and materials, 57 (2012) 1, 135-143.

[9] M. Warzecha, T. Merder: Numerical analysis of the non-metallic, inclusions distribution and separation in a two-strand tundish, Metalurgija - Journal for Theory and Prac-tice in Metallurgy, 52 (2013) 2, 153-156.

[10] K. Michalek, K. Gryc, J. Morávka: Physical modellign of bath homogenisation in argon stirred ladle. Metalurgija, 48 (2009) 4, 215-218.

[11] K. Michalek, Z. Hudzieczek, K. Gryc, Mathematical iden-tifi cation of homogenisation processes in argon stirred lad-le, Metalurgija, 48 (2009) 4, 219-222.

[12] T. Merder: Effect of casting fl ow rate on steel fl ow pheno-mena in tundish, Metalurgija - Journal for Theory and Practice in Metellurgy, 52 (2013) 2, 161-164 .

[13] M. Warzecha: Modeling of the steel fl ow and non-metallic inclusion separation due to fl otation in a six-strand cc tun-dish, Materials Science Forum. 706-709 (2012) 1556-1561.

[14] J. Bažan, Z. Bužek, J. Roučka, K. Strańsky, P. Lev: O me-chanizmu fi ltrace tvarne między ceditkovymi a pénovymi fi ltry. Mat. VIII Międzynarodowej Konferencji Naukowej „Iron and Steelmaking”, Mala Lučivna, 23-25.09.1998, 168-171

Note: The responsible translator for English language is Z. Kozyra, Katowice, Poland

industrial investigations of the liquid steel filtration

Documents